Open Evidence, in consortium with RAND Europe, conducted the study “Media literacy and online empowerment issues raised by algorithm-driven media services” for DG CONNECT, which has been published and is now available at the European Commission’s website. The study’s objectives were to explore how the provision and consumption of news in social media platforms function, and to identify which problems arise and potentially hinder the functioning of democracies. The study also analysed initiatives that are currently in place to counter these issues and their implications. Finally, it also provided other options to be explored. The research approach included a set of quantitative and qualitative methodologies, including a review of the literature, key informant interviews, statistical analysis, a mapping of stakeholders and interventions, and assessment of the feasibility of using behavioural experiments in this area.
In the past few years, there has been growing concern about the impact of algorithm-driven media services and their potential role in the political and democratic landscape. Among the existing solutions, several actions aiming at the empowerment of the individuals were detected. While these solutions do not directly address the root causes of disinformation or the quality of information in circulation, they aim to minimise the impact of user exposure to disinformation by increasing the confidence, skills and knowledge of users of platforms to effectively navigate the modern media environment and critically assess the media to which they are exposed. The study also identified knowledge gaps in the understanding of and the available research on the algorithm-driven media services, the role of such services in shaping socio-political discourse, and their effect on wider societal outcomes.
The study showed that users tend to be unaware of their own cognitive biases. In this regard, Open Evidence used its expertise in behavioural sciences to explore the potential of reinforcing solutions, including awareness-raising and media literacy interventions. The study proposed “media literacy by design”, a concept that captures the need to embed media literacy into de choice architecture, triggering more information and media literate use. To this end, the study proposed three experiments that could be conducted to assess the effect of the type of interventions on the online behaviour of social media users.